Previous epidemiological research on individual dietary components and risk for Alzheimer's disease (AD) has produced conflicting results. Although individual foods or nutrients are not consumed in isolation but rather as components of the daily diet, the effect of composite dietary patterns has not been explored in the AD literature. Defining diet by dietary patterns has the ability to capture its multidimensionality while reducing its apparent complexity because patterns can integrate complex or subtle interactive effects of many dietary constituents and bypass problems generated by multiple testing and the high correlations that may exist among these constituents. A dietary pattern example is the Mediterranean diet (MeDi), which has received increased attention in recent years because of converging evidence that it is related to lower risk for cardiovascular disease, several forms of cancer and overall mortality. The MeDi seems to include many of the components reported as potentially beneficial for AD. Preliminary exploration of our dietary data from a longitudinal population study of elderly subjects in New York indicated that higher adherence to the MeDi is related to decreased risk for development of AD. We now want to 1) investigate possible biological mechanisms of this association and 2) explore the existence of other than the MeDi dietary patterns that could also confer protection for AD risk. We intend to use biomarkers involved in oxidative (isoprostane 8,12-iso-Ipf2a-VI), inflammatory (C-reactive protein, interleukin-6, interleukin-1), metabolic (insulin, adiponectin) and vascular (homocysteine, MRI white matter hyperintensities and MRI subclinical infarcts) pathways in order to gain further insight into the mechanisms by which the MeDi exerts its biological effect for AD protection. Because our population in Manhattan is not the paradigmatic one in terms of MeDi adherence, it is possible that combinations of foods or nutrients, other than the one describing the MeDi, could better fit its specific dietary habits. We plan to use reduced rank regression and canonical variance analyses, 2 modern statistical methods that combine a priori and a posteriori approaches, in order to explore which particular food components of the diet carry most of the protective for AD effect and define other dietary patterns that could be protective for AD, conceivably even more than the MeDi. The American Heart Association has issued a scientific advisory stating that the Mediterranean Diet demonstrates impressive effects on cardiovascular disease and the magnitude of the effect of the Mediterranean Diet for Alzheimer's disease protection seen in our preliminary studies (40% risk reduction for the higher adherence tertile) was similar to effects seen for cardiovascular disease. Additionally, dietary interventions have been often shown to be even more efficacious than medications. Therefore, elucidating the biological mechanisms of the Mediterranean Diet's protection for Alzheimer's disease (a disease that will be affecting 14 million Americans by 2050, with no known ways to prevent, delay or treat it) is of major importance, from a public health perspective.